Pediatric Endoscopic Gastrostomy Tubes

Abstract

See “Long-Term Outcomes of Infants and Children Undergoing Percutaneous Endoscopy Gastrostomy Tube Placement” by McSweeney et al in the November 2013 issue (JPGN 57:663-7). Placement of percutaneous endoscopic gastrostomy (PEG) tubes has been routinely practiced in pediatric medical centers since 1980, when the procedure was introduced to replace the invasive, open surgical approach (1). Several studies describe short-term PEG complication rates in pediatric patients that range from 4% to 50% (2–6). These complications have included death, perforation, peritonitis, fistula, infection, gastroesophageal reflux, major granulation tissue, or buried bumpers. A few smaller studies have reported long-term cumulative complication rates of 44% at 1 to 8 years and 47.7% at 24 months of age (6,7). A recent report noted a major complication rate of 18.3%, including buried or extruded buttons, granulation, fistula, wound infection, gastrostomy closure delay, peritonitis, neuralgia, nonprophylactic antibiotics, blood transfusions, and contingencies requiring surgical revisions or endoscopic procedures (8). No deaths were associated with the procedure. In the November 2013 issue of the Journal of Pediatric Gastroenterology and Nutrition, McSweeney et al (9) reported a study focused on complication rates and need for a PEG during a 10-year follow-up period, based on a retrospective review of PEGs installed between 1999 and 2000 (21 months). For purposes of the study, the authors defined a major complication as additional hospitalization or any surgical or interventional radiology procedures. This study was done well and supplies important long-term data for proceduralists and surgeons counseling families for a PEG. The most significant contribution and major strength of this retrospective study are the relatively long follow-up period covered. A cumulative complication rate of 15% was found at 5.4 years. PEG tubes were used for a median of 10.2 years. Some limitations of the study were the single-center design and potential bias in evaluating performance at one's own institution. The outcomes may reflect local factors, including patient selection criteria for PEG placement, postoperative and outpatient care, training, and skill of the operators. PEG placement in this study also required a gastroenterologist to visualize the stoma site and a surgeon to perform the puncture. The involvement of 2 operators may have led to different long-term complication rates. Going forward, newer techniques such as laparoscopy-assisted PEG placement may also change long-term complications rates. A further concern is the definition of major complications (adapted from an article at the same institution) (10). This protocol prompted the exclusion of significant infections treated with antibiotics on an outpatient basis, granulation requiring cauterization and/or medical therapy, and nerve injury, a category addressed by these authors in earlier articles. For example, McSweeney et al noted that cellulitis comprises 66% of the major complications that also required hospitalization; a pediatric population with 56.5% neurologic impairment and 23.9% metabolic or genetic disorders is often hospitalized for indications other than a gastrostomy tube cellulitis that may have been treated as part of such an admission. These considerations raise the possibility of both underreporting and overreporting complications. Finally, outcomes data may be skewed by subjects lost to follow-up (14%). An important aspect of this article, albeit not its primary intent, was the duration of tube placement. In this group of approximately 80% neurologically impaired and genetically abnormal children, the median placement time was 10.2 years, with a 65% retention rate at 5 years. Whether these results apply to children without neurologic or genetic abnormalities remains to be determined, but the results do represent a starting point for a discussion about length of PEG tube need. In summary, McSweeney et al offer important observations on long-term complications of PEG placement in a select group of children and adolescents. The results should be interpreted with caution, given the single-center design, restrictive operational definition of complications, and patient population studied. As the authors suggest, further analysis of the cost of tube placement in various ways, tube retention by patient population, and evaluation of specific complications may further improve the quality of information available to guide decisions regarding the placement of PEGs. Decisions surrounding enteral nutrition for a child can be emotional for parents. Clinicians need reliable, unbiased evidence to present to families if they are to achieve the goal of shared decision making. This article brings important information to the field regarding such goals, but more research is needed. A national consensus on outcomes of PEG procedures and other types of gastrostomy tube insertions would further assist physicians and parents in making informed decisions.